Server cooling system, server fan speed regulation method and baseboard management controller

Abstract

The present application discloses a server cooling system, a server fan speed regulation method, and a baseboard management controller, relating to the technical field of servers. By separately providing an electronic fuse unit for each fan in a server fan board, the baseboard management controller can determine the power consumption value of each fan based on the fan current and voltage values ​​collected by the electronic fuse unit, thereby determining the energy efficiency value of the server based on the average total power consumption of the fans and the average power consumption of the server in two adjacent cycles. Furthermore, the speed of each fan can be regulated based on the energy efficiency values ​​of the two adjacent cycles, thereby achieving performance speed regulation of the server and enabling the server to operate at better energy efficiency.

Description

Technical Field

[0001] The present application relates to the field of server technology, and in particular to a server heat dissipation system, a server fan speed regulation method, and a baseboard management controller. Background Art

[0002] To ensure stable server operation, the server needs to be cooled. Air cooling is a common method of cooling a server, which uses fans to remove heat from the server.

[0003] Related technologies typically use a combination of inlet temperature and proportional-integral-derivative (PID) speed control to adjust fan speeds. However, this approach struggles to meet server energy efficiency requirements. Summary of the Invention

[0004] The present application provides a server heat dissipation system, a server fan speed regulation method and a baseboard management controller, so as to at least solve the problem in the related art that it is difficult to meet the energy efficiency requirements of the server.

[0005] The present application provides a server cooling system, comprising: a baseboard management controller, a fan board, and a plurality of electronic fuse units; a plurality of fans are provided on the fan board, each fan is electrically connected to a corresponding electronic fuse unit, and each electronic fuse unit is electrically connected to the baseboard management controller;

[0006] The electronic fuse unit is used to collect the current and voltage values ​​of the fan at each sampling moment in the current cycle and send them to the baseboard management controller;

[0007] The baseboard management controller is used to determine the average total power consumption of the fans in the current cycle based on the current value and voltage value of each fan at each sampling moment in the current cycle;

[0008] The baseboard management controller is further used to determine the energy efficiency value of the server according to the average total power consumption of the fans and the average power consumption of the server in two adjacent cycles, and to adjust the speed of each fan according to the energy efficiency value of the two adjacent cycles.

[0009] This application provides a server fan speed adjustment method, including:

[0010] Obtaining the initial fan speed value corresponding to the current inlet air temperature, the average power consumption of the first server in the previous period, and the average total power consumption of the first fan;

[0011] If the working state of the server in the previous cycle meets the preset working state condition, a first energy efficiency value of the server in the previous cycle is determined according to the first server average power consumption and the first fan average total power consumption, and the efficiency speed regulation logic is triggered;

[0012] The efficiency speed control logic includes:

[0013] Controlling each fan in the server to rotate according to a first fan speed value in a current cycle; the first fan speed value is less than an initial fan speed value;

[0014] Determine a second energy efficiency value of the server in the current period according to the second server average power consumption and the second fan average total power consumption in the current period;

[0015] The rotation speed of each fan is dynamically adjusted according to the first energy efficiency value and the second energy efficiency value.

[0016] The present application also provides a server fan speed regulating device, comprising:

[0017] An acquisition module is used to obtain an initial fan speed value corresponding to the current inlet air temperature, an average power consumption of the first server in the previous period, and an average total power consumption of the first fan;

[0018] a determination module configured to determine a first energy efficiency value of the server in the previous cycle based on the first server average power consumption and the first fan average total power consumption, and trigger an efficiency speed regulation logic if the server's operating status in the previous cycle meets a preset operating status condition;

[0019] Efficiency speed control module, used for:

[0020] Controlling each fan in the server to rotate according to a first fan speed value in a current cycle; the first fan speed value is less than an initial fan speed value;

[0021] Determine a second energy efficiency value of the server in the current period according to the second server average power consumption and the second fan average total power consumption in the current period;

[0022] The rotation speed of each fan is dynamically adjusted according to the first energy efficiency value and the second energy efficiency value.

[0023] The present application also provides a baseboard management controller, comprising: a memory for storing a computer program; and a processor for implementing the steps of any of the above-mentioned server fan speed regulation methods when executing the computer program.

[0024] The present application also provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned server fan speed regulation methods are implemented.

[0025] The present application also provides a computer program product, including a computer program, which implements the steps of any of the above-mentioned server fan speed regulation methods when executed by a processor.

[0026] The present application sets an electronic fuse unit for each fan in the server fan board. The baseboard management controller can determine the power consumption value of each fan based on the current and voltage values ​​of the fan collected by the electronic fuse unit, thereby determining the energy efficiency value of the server based on the average total power consumption of the fans and the average power consumption of the server in two adjacent cycles. The speed of each fan can then be adjusted according to the energy efficiency values ​​of the two adjacent cycles, thereby achieving performance speed regulation of the server and enabling the server to operate at better energy efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0028] Figure 1 A schematic diagram of the structure of a server cooling system provided in an embodiment of the present application;

[0029] Figure 2 Schematic diagram of the process of the server fan speed adjustment method provided in the embodiment of the application Figure 1 ;

[0030] Figure 3 Schematic diagram of the process of the server fan speed adjustment method provided in the embodiment of the application Figure 2 ;

[0031] Figure 4 Schematic diagram of the process of the server fan speed adjustment method provided in the embodiment of the application Figure 3 ;

[0032] Figure 5 A flowchart of a fault determination process provided in an embodiment of the present application;

[0033] Figure 6 A schematic diagram of the structure of a server fan speed regulating device provided in an embodiment of the present application;

[0034] Figure 7 A schematic diagram of the structure of the baseboard management controller provided in an embodiment of the present application. DETAILED DESCRIPTION

[0035] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0036] It should be noted that, in the description of this application, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. The terms "first," "second," etc., in this application are used to distinguish similar objects, and are not used to describe a particular order or sequence.

[0037] With the continuous development of the 5G (fifth-generation mobile communication technology) era, artificial intelligence, industrial Internet, and cloud computing industries, more and more data centers are being established. Servers are the core equipment in data centers, and performance, availability, cost-effectiveness, and efficiency have become important indicators for measuring the quality of servers.

[0038] As the power consumption of key components continues to rise, fan performance is constantly improving to address internal server component heat dissipation. This, in turn, increases fan speed and power consumption. As fan power consumption increases, server energy efficiency also varies significantly depending on the configuration. For example, key components include the central processing unit (CPU), memory, graphics processing unit (GPU), and other high-power components.

[0039] Related technologies typically use a combination of inlet temperature control and PID speed control. The baseboard management controller (BMC) monitors the temperatures of key components such as the CPU, memory, and GPU in real time, and uses PID speed control to calculate the required speed of each key component to adjust the fan speed. The PID speed control calculation formula is as follows:

[0040] pwm(t)=pwm(t-1)+pwm_vary(t)

[0041]

[0042] Where pwm(t) is the fan speed currently requested by the component, pwm(t-1) is the fan speed requested by the component in the previous second, pwm_vary(t) is the fan speed variation, dt=1 indicates a 1-second frequency, Kp is the P value in the PID speed control parameter, Ki is the I value in the PID speed control parameter, Kd is the D value in the PID speed control parameter, T(t) is the current component temperature, T(t-1) is the component temperature in the previous second, T(t-2) is the component temperature in the previous two seconds, and SP is the component temperature control point.

[0043] However, the above speed regulation method is difficult to meet the energy efficiency requirements of the server.

[0044] Based on this, the server cooling system and server fan speed regulation method provided by the present application, by separately setting an electronic fuse unit for each fan in the server fan board, the baseboard management controller can determine the power consumption value of each fan according to the current value and voltage value of the fan collected by the electronic fuse unit, thereby being able to determine the energy efficiency value of the server according to the average total power consumption of the fan and the average power consumption of the server in two adjacent cycles, and then being able to regulate the speed of each fan according to the energy efficiency value of the two adjacent cycles, thereby realizing the efficiency speed regulation of the server, so that the server can operate at better energy efficiency.

[0045] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0046] Figure 1 This is a schematic diagram of the structure of the server cooling system provided in the embodiment of the present application. Figure 1 The server cooling system includes a baseboard management controller 10, a fan board 20 and multiple electronic fuse units 30; multiple fans 40 are provided on the fan board 20, each fan 40 is electrically connected to a corresponding electronic fuse unit 30, and each electronic fuse unit 30 is electrically connected to the baseboard management controller 10.

[0047] The total number of the multiple fans 40 provided on the fan plate 20 can be set according to actual needs, which is not limited in this application, such as 2, 4, 6, 8 or 12. Accordingly, a corresponding electronic fuse unit 30 is provided for each fan 40, so the total number of the multiple electronic fuse units 30 is the same as the total number of the multiple fans 40. For ease of representation, Figure 1 Only two fans 40 and two electronic fuse units 30 are taken as an example.

[0048] The electronic fuse unit 30 is used to collect the current value and voltage value of the fan 40 at each sampling moment in the current cycle and send them to the baseboard management controller 10;

[0049] The baseboard management controller 10 is configured to determine the average total power consumption of the fans in the current cycle based on the current value and voltage value of each fan 40 at each sampling moment in the current cycle;

[0050] The baseboard management controller 10 is further configured to determine the energy efficiency value of the server according to the average total power consumption of the fans and the average power consumption of the server in two adjacent cycles, and to adjust the speed of each fan 40 according to the energy efficiency value in two adjacent cycles.

[0051] By separately providing an electronic fuse unit 30 for each fan 40 in the fan board 20 of the server, the baseboard management controller 10 can determine the power consumption value of each fan 40 based on the current value and voltage value of the fan 40 collected by the electronic fuse unit 30, thereby determining the energy efficiency value of the server based on the average total power consumption of the fans and the average power consumption of the server in two adjacent cycles. Furthermore, the speed of each fan 40 can be adjusted based on the energy efficiency value of the two adjacent cycles, thereby achieving performance speed regulation of the server and enabling the server to operate at better energy efficiency.

[0052] The electronic fuse (eFuse) unit 30 uses a built-in current sensor to monitor the fan branch current (I_fan) in real time. When the current exceeds a preset threshold, the eFuse unit 30 triggers a protection mechanism (such as power cutoff) to prevent damage to the fan or circuit. The preset threshold can be set based on actual needs and is not limited in this application.

[0053] The electronic fuse unit 30 is also capable of measuring the voltage (V_fan) at the output of the fan 40 .

[0054] At each sampling moment, the baseboard management controller 10 is used to calculate the power consumption value of the fan 40 at the sampling moment according to the current value and voltage value of the fan 40 at the sampling moment according to the power consumption calculation formula. The power consumption calculation formula is as follows:

[0055] P_fan = V_fan × I_fan

[0056] Wherein, P_fan is the power consumption value of the fan 40 .

[0057] For any cycle, the baseboard management controller 10 is used to determine the sum of the power consumption values ​​of each fan 40 in the fan board 20 at each sampling moment as the total fan power consumption of the cycle, and determine the ratio of the total fan power consumption to the total number of multiple fans 40 as the average total fan power consumption of the cycle.

[0058] The baseboard management controller 10 determines the energy efficiency value of the server based on the average total power consumption of the fans and the average power consumption of the servers in two adjacent cycles, and adjusts the speed of each fan 40 based on the energy efficiency values ​​of the two adjacent cycles. The specific implementation process is shown in the following method embodiment and is not repeated here.

[0059] In one possible implementation, the electronic fuse unit 30 is further configured to monitor the status of the fan 40 and transmit the status information of the fan 40 to the baseboard management controller 10 via a digital interface (e.g., I²C). The status information indicates the status of the fan 40, such as whether it is powered on or has an overcurrent.

[0060] The baseboard management controller 10 monitors the status of the fans 40 in real time through the electronic fuse unit 30 and obtains the power consumption value and status information of each fan 40. Therefore, the baseboard management controller 10 can quickly identify the faulty fan based on the status information fed back by the electronic fuse unit 30.

[0061] Since the electronic fuse unit 30 can also collect the current and voltage values ​​of the fan 40 , the baseboard management controller 10 can determine the power consumption value of the fan 40 at each moment. Therefore, the baseboard management controller 10 can also analyze the cause of the fault based on the power consumption value of the faulty fan.

[0062] Accordingly, the baseboard management controller 10 is further configured to:

[0063] When any fan 40 fails, the power consumption value of the failed fan is compared with the power consumption values ​​of other fans in the plurality of fans 40 except the failed fan;

[0064] If the power consumption value of the faulty fan is not zero and the difference between the power consumption value of the faulty fan and any other fan is less than the preset difference, the fault is determined to be caused by an abnormal frequency generator signal (FG); or

[0065] If the power consumption value of the faulty fan is not zero and the difference between the power consumption values ​​of the other fans is greater than or equal to the preset difference, the fault is determined to be caused by an abnormal pulse width modulation (PWM) signal; or

[0066] If the power consumption of the faulty fan is 0, the fault is determined to be caused by an abnormal power supply signal, such as a power failure.

[0067] By capturing the precise power consumption value of the faulty fan through the BMC and comparing it with the power consumption values ​​of other fans, it is possible to accurately determine whether the faulty fan is caused by poor signal contact or a fault in the fan itself, thereby quickly locating the specific cause of the fan failure.

[0068] As the power consumption of key components in the server increases, the temperature changes and fan speed requirements brought about by the increase in power consumption when the key components are pressurized are also increasing. If the speed value in the inlet air temperature speed control curve is set too low, the speed increase rate will not keep up with the fan speed required by the component, and there will be a risk of component overheating.

[0069] In a possible implementation, the baseboard management controller 10 is further configured to:

[0070] When the speed of each fan 40 is adjusted according to the energy efficiency values ​​of two adjacent cycles, the high-parameter inlet air temperature speed adjustment curve is used for speed adjustment.

[0071] The above implementation method can be combined with the performance speed regulation function to gradually reduce the fan speed when there is no sudden pressure on the power consumption of key components, so as to achieve the purpose of achieving better performance and lower noise for the server. When setting the parameters of the inlet temperature speed regulation curve, the speed value of the entire curve can be set to a higher value, for example, the speed value can be uniformly increased by 10%, 20% or 30%. The specific setting can be based on actual needs, and this application does not limit this. In this way, when high-power consumption components such as the CPU are stressed or the power consumption is suddenly increased, the fan speed can be quickly increased, effectively alleviating the overshoot temperature and overheating phenomenon when the power consumption of high-power consumption components such as the CPU is suddenly increased, so that the server can operate in a state of better performance and lower noise.

[0072] The server cooling system provided in this application can be used for air-cooled servers of different models (such as 1U, 2U, 3U, etc.).

[0073] The execution subject of the server fan speed regulation method provided in the present application is a server fan speed regulation device, which is integrated into the baseboard management controller 10.

[0074] Figure 2 Schematic diagram of the process of the server fan speed adjustment method provided in the embodiment of the application Figure 1 ,like Figure 2 As shown, the server fan speed adjustment method includes:

[0075] S201: Obtain an initial fan speed value corresponding to the current air inlet temperature, an average power consumption of the first server in the previous period, and an average total power consumption of the first fan.

[0076] The fan speed value corresponding to the inlet air temperature can be obtained according to a preset inlet air temperature speed control curve. The inlet air temperature speed control curve includes the fan speed values ​​corresponding to the inlet air temperature at each moment.

[0077] The first server average power consumption is an average value of the server power consumption at each sampling moment in the previous cycle.

[0078] The first fan's average total power consumption is the average of the sum of the fan power consumption values ​​at each sampling moment in the previous cycle. The fan power consumption sum at any sampling moment is the sum of the power consumption values ​​of multiple fans at that sampling moment.

[0079] The interval between two adjacent sampling moments can be set according to actual needs, and this application does not limit this, for example, 1s, 2s, etc., that is, the sampling frequency is 1s / time, 2s / time, etc.

[0080] The duration of each cycle can be set according to actual needs, and this application does not limit it, for example, 5 minutes, 10 minutes, etc.

[0081] S202: If the working state of the server in the previous cycle meets the preset working state condition, determine the first energy efficiency value of the server in the previous cycle according to the first server average power consumption and the first fan average total power consumption, and trigger the efficiency speed regulation logic.

[0082] If the server's operating status during the previous cycle meets the preset operating conditions, it indicates that the server has been operating relatively stably during this period, with no sudden high load or temperature anomalies. At this point, you can try performance adjustment to optimize the server's energy efficiency.

[0083] The first energy efficiency value (η0) is used to represent the energy utilization efficiency of the server in the previous cycle.

[0084] The higher the energy efficiency value, the more work the server completes while consuming the same amount of energy, and the more efficient the energy utilization.

[0085] The performance speed regulation logic includes S203-S205.

[0086] S203: Control each fan in the server to rotate according to a first fan speed value in the current cycle; the first fan speed value is smaller than the initial fan speed value.

[0087] By reducing the fan speed, we attempt to reduce fan energy consumption and improve server energy efficiency while ensuring server heat dissipation. This is a tentative adjustment, and further adjustments to the fan speed will be made based on changes in energy efficiency and other conditions.

[0088] S204: Determine a second energy efficiency value of the server in the current period according to the second server average power consumption and the second fan average total power consumption in the current period.

[0089] The second energy efficiency value (η1) is used to represent the energy utilization efficiency of the server in the current cycle.

[0090] S205: Dynamically adjust the rotation speed of each fan according to the first energy efficiency value and the second energy efficiency value.

[0091] Exemplarily, by comparing the second energy efficiency value η1 of the current cycle with the first energy efficiency value η0 of the previous cycle, it is determined whether the fan speed adjustment is effective, thereby dynamically adjusting the speed of each fan.

[0092] This embodiment adjusts the speed of each fan according to the energy efficiency values ​​of two adjacent cycles, thereby achieving performance speed adjustment of the server, so that the server can operate at better energy efficiency.

[0093] exist Figure 2 Based on the illustrated embodiment, a server fan speed adjustment method is described in detail below using a specific embodiment.

[0094] Figure 3 Schematic diagram of the process of the server fan speed adjustment method provided in the embodiment of the application Figure 2 ,like Figure 3 As shown, the server fan speed adjustment method includes:

[0095] S301: Obtain the average power consumption of the first server in the previous period.

[0096] For example, the BMC reads the server power consumption every second, accumulates the data for 10 minutes (600 seconds), and takes the average power consumption value over 10 minutes as the average server power consumption (STP) for the cycle. This eliminates short-term power consumption fluctuations and produces a relatively stable server power consumption value. This method obtains the average power consumption of the first server in the previous cycle.

[0097] S302: Obtain the component power consumption value of the key component at each sampling moment in the previous cycle, and determine the power consumption fluctuation value according to the component power consumption values ​​at two adjacent sampling moments.

[0098] For example, key components include CPU, memory, GPU, or other high-power consumption components.

[0099] For example, the BMC reads the power consumption of key components every second. The current component power consumption value is CP0, and the component power consumption value read the next second is CP1. The power consumption fluctuation value ∆CP is calculated as: ∆CP = (CPU1 - CP0) / CP0.

[0100] High-power components like the CPU and GPU generate the most heat in a server. Changes in their power consumption directly reflect changes in workload. By monitoring the power consumption of these components in real time, you can keep abreast of the server's operating status.

[0101] S303: Obtain the average total power consumption of the first fan in the previous cycle.

[0102] For example, the BMC reads the power consumption of all fans on the fan board every second, recording the total power consumption for 10 minutes (600 seconds). The 10-minute average power consumption is taken as the average total fan power consumption (FTP) for the cycle. This eliminates short-term fluctuations in fan power consumption and produces a stable total fan power consumption value. This method obtains the average total power consumption of the first fan in the previous cycle.

[0103] S304: Obtaining an initial fan speed value corresponding to the current air inlet temperature.

[0104] In one possible implementation, a high-parameter inlet air temperature speed control curve is used to obtain a fan speed value corresponding to the inlet air temperature. Accordingly, the specific implementation process of obtaining the initial fan speed value corresponding to the current inlet air temperature includes: obtaining a preset high-parameter inlet air temperature speed control curve; and obtaining the initial fan speed value corresponding to the current inlet air temperature from the inlet air temperature speed control curve.

[0105] The inlet air temperature speed control curve is developed based on the server's cooling design and experimental data. It establishes a corresponding relationship between inlet air temperature and fan speed. By querying the position of the current inlet air temperature T0 on the inlet air temperature speed control curve, you can obtain the corresponding fan speed value FDT0, which is the basic fan speed setting for the server at the current ambient temperature.

[0106] By obtaining the initial fan speed value corresponding to the current air inlet temperature, data support is provided for subsequent performance speed regulation.

[0107] This application does not limit the execution order of S301, S302, S303 and S304. For example, S301, S302, S303 and S304 can be executed in sequence or simultaneously. Figure 3 An example is given by sequentially executing S301 , S302 , S303 and S304 .

[0108] S305: If the working state of the server in the previous cycle meets the preset working state condition, determine the first energy efficiency value of the server in the previous cycle according to the first server average power consumption and the first fan average total power consumption, and trigger the efficiency speed regulation logic.

[0109] In one possible implementation, the specific implementation process of determining whether the working status of the server in the previous cycle meets the preset working status condition includes:

[0110] Obtain the speed requirements of key components in the server;

[0111] Compare the speed demand value of the key component with the initial fan speed value to determine whether the key component triggered the proportional integral derivative speed regulation in the previous cycle;

[0112] If the key components did not trigger the proportional integral differential speed control in the previous cycle, and the speed of each fan was controlled according to the inlet air temperature speed control curve, then it is determined that the working status of the server in the previous cycle meets the preset working status conditions; otherwise, it is determined that the working status of the server in the previous cycle does not meet the preset working status conditions.

[0113] For example, if no component triggers PID speed regulation for 10 consecutive minutes, and the fan speed is controlled according to the inlet air temperature speed regulation curve, it indicates that the server is in a relatively stable operating state during this period, without sudden high load or temperature anomalies. In this case, the server can be considered to meet the preset operating conditions. Conversely, if a key component triggers PID speed regulation, it indicates that the server's operating state is unstable during this period, possibly with sudden high load or temperature anomalies. In this case, the server can be considered to not meet the preset operating conditions.

[0114] The condition for triggering PID speed regulation is: comparing the speed demand value FDX of the key component with the initial fan speed value FDT0. If FDX ≥ FDT0, PID speed regulation is triggered.

[0115] If the server is in a relatively stable working state in the previous cycle and there is no sudden high load or temperature abnormality, performance speed regulation can be attempted to optimize the server's energy efficiency. Therefore, by judging whether the server's working state in the previous cycle meets the preset working state conditions, it provides a trigger condition for the subsequent performance speed regulation logic.

[0116] In a possible implementation, the specific implementation process of determining the first energy efficiency value of the server in the previous period based on the average power consumption of the first server and the average total power consumption of the first fan in S305 includes: determining the difference between the first preset value and the energy efficiency ratio as the first energy efficiency value.

[0117] The first preset value may be 1.

[0118] The energy efficiency ratio is the ratio of the average total power consumption of the first fan to the average power consumption of the first server.

[0119] This implementation provides a method for calculating an energy efficiency value, by calculating the ratio of the average total power consumption of the first fan to the average total power consumption of the first server and subtracting the ratio from 1 to obtain a relatively accurate first energy efficiency value.

[0120] The performance speed regulation logic in this embodiment includes S306 - S308 .

[0121] S306: Control each fan in the server to rotate according to a first fan speed value in the current cycle.

[0122] In a possible implementation, before S306 , the first fan speed value is calculated. Accordingly, the method provided in the present application further includes: determining the difference between the initial fan speed value and the second preset value as the first fan speed value.

[0123] The second preset value may be 1.

[0124] The fan duty is reduced by 1 unit (FD1 = FDT0 - 1) in an attempt to reduce fan energy consumption and improve server energy efficiency while ensuring server heat dissipation.

[0125] This is a tentative adjustment, and the subsequent decision on whether to continue adjusting the fan speed will be based on changes in energy efficiency values ​​and other conditions.

[0126] S307: Determine a second energy efficiency value of the server in the current period according to the second server average power consumption and the second fan average total power consumption in the current period.

[0127] Exemplarily, the difference between the first preset value and the energy efficiency ratio is determined as the second energy efficiency value; wherein the energy efficiency ratio is the ratio of the average total power consumption of the second fan to the average power consumption of the second server.

[0128] The first preset value may be 1.

[0129] S308: Dynamically adjust the rotation speed of each fan according to the first energy efficiency value and the second energy efficiency value.

[0130] In a possible implementation, when the second energy efficiency value is greater than or equal to the first energy efficiency value, the specific implementation process of S308 includes the following four cases.

[0131] If η1 ≥ η0, it means that the energy efficiency of the server is improved after reducing the fan speed, and you can continue to try to reduce the fan speed.

[0132] First, determine whether the key component triggers PID speed regulation in the current cycle. Specifically, obtain the speed demand value of the key component; compare the speed demand value of the key component with the first fan speed value to determine whether the key component triggers PID speed regulation in the current cycle.

[0133] The first case: If the key component does not trigger the proportional integral differential speed regulation in the current cycle, and the power consumption fluctuation value of the key component is less than the preset fluctuation value, the difference between the first fan speed value and the second preset value is determined as the second fan speed value, and each fan is controlled to rotate according to the second fan speed value.

[0134] The preset fluctuation value can be 30% or other fluctuation values ​​set according to actual needs, and this application does not limit this.

[0135] The second preset value may be 1.

[0136] In this case, the server's operating status is relatively stable, with no sudden high load or abnormal temperature. The fan speed can be further reduced (FD2=FD1-1) to further optimize energy efficiency.

[0137] The second case: If the key component triggers proportional integral differential speed regulation in the current cycle, and the power consumption fluctuation value of the key component is less than the preset fluctuation value, the performance speed regulation logic is interrupted, and the fan demand value corresponding to the key component is determined as the second fan speed value, and each fan is controlled to rotate according to the second fan speed value.

[0138] When the PID speed control is triggered, it means that the workload of the component has changed and the corresponding fan speed needs to be adjusted. At this time, the performance speed control is interrupted and the fan speed is set to the fan speed FDX required by the component to ensure the cooling requirements of the component.

[0139] The third case: If the key component does not trigger the proportional integral differential speed control in the current cycle, and the power consumption fluctuation value of the key component is greater than or equal to the preset fluctuation value, the performance speed control logic is interrupted, and the fan speed value corresponding to the current air inlet temperature is determined as the second fan speed value, and each fan is controlled to rotate according to the second fan speed value.

[0140] A significant change in the power consumption of key components (∆CP ≥ 30%) indicates that the server's operating state is about to undergo a significant change. At this point, performance speed regulation is interrupted and the fan speed is reset to the fan speed value FDTX corresponding to the current inlet air temperature to address the potential increase in cooling demand.

[0141] The fourth case: If the key component triggers proportional integral differential speed regulation in the current cycle, and the power consumption fluctuation value of the key component is greater than or equal to the preset fluctuation value, the performance speed regulation logic is interrupted, and the maximum value of the fan demand value corresponding to the key component and the fan speed value corresponding to the current inlet air temperature is determined as the second fan speed value, and each fan is controlled to rotate according to the second fan speed value.

[0142] In this situation, the server's operating status experiences both component workload variations and significant power consumption fluctuations in key components. To ensure safe cooling, performance speed regulation is interrupted and the fan speed FDX required by the component is compared with the fan speed FDTX corresponding to the current inlet air temperature. The higher of the two is used as fan speed FD2 to provide sufficient cooling capacity.

[0143] In a possible implementation, when the second energy efficiency value is less than the first energy efficiency value, the specific implementation process of S308 includes: controlling each fan to continue rotating according to the first fan speed value.

[0144] If η1<η0, it means that reducing the fan speed will cause the server energy efficiency to decrease, and the current fan speed needs to be maintained.

[0145] By repeatedly repeating the aforementioned efficiency-adjusted speed logic and calculations, the fan speed is dynamically adjusted based on the server's operating status and energy efficiency changes, ensuring optimal energy efficiency while ensuring heat dissipation. This dynamic adjustment mechanism optimizes fan speed in real time based on the server's actual operating conditions, improving energy efficiency and reducing operating costs.

[0146] Based on the above embodiments, Figure 4 Schematic diagram of the process of the server fan speed adjustment method provided in the embodiment of the application Figure 3 See also Figure 4 First, for each cycle, the BMC collects power consumption information for key components (such as the CPU, GPU, or other high-power components), server power consumption, total fan power consumption, and temperature information. Temperature information refers to the current inlet air temperature, which is used to obtain the fan speed value corresponding to the current inlet air temperature based on the inlet air temperature speed control curve. If the current fan speed is FDT0 and the fan speed continues to follow the inlet air temperature speed control curve for 10 minutes, the fan speed value for the next cycle, FD1, is determined as TDT0-1. Based on FD1, the initial energy efficiency value η0, the energy efficiency value η1 for the next cycle, and the speed control status and power consumption status monitoring, the fan speed value FD2 for the next cycle is determined. The current cycle, the next cycle, and the next cycle are three consecutive cycles.

[0147] Based on the above embodiments, Figure 5 This is a flowchart of the fault determination process provided in the embodiment of the present application. Figure 5 , the BMC monitors the status of the fans in real time and obtains the power consumption value and status information of each fan, so that the faulty fan can be quickly identified based on the status information. The BMC can also analyze the cause of the fault based on the power consumption value of the faulty fan. Accordingly, the method provided by this application also includes:

[0148] When any fan fails, the power consumption value of the failed fan is compared with the power consumption values ​​of other fans except the failed fan among the multiple fans in the server;

[0149] If the power consumption value of the faulty fan is not 0 and the difference between the power consumption value of the faulty fan and any other fan is less than the preset difference, the fault is determined to be caused by abnormal FG signal; or

[0150] If the power consumption value of the faulty fan is not 0 and the difference between the power consumption values ​​of the other fans is greater than or equal to the preset difference, it is determined that the fault is caused by abnormal PWM signal; or,

[0151] If the power consumption of the faulty fan is 0, the fault is determined to be caused by an abnormal power supply signal.

[0152] By capturing the precise power consumption value of the faulty fan through the BMC and comparing it with the power consumption values ​​of other fans, it is possible to accurately determine whether the fan fault is caused by poor signal contact or the fan itself, thereby quickly locating the specific cause of the fan fault.

[0153] Continue to see Figure 5 After determining the cause of a fault, the BMC can also trigger an alarm. For example, it can output an alarm message to indicate the fan fault and the cause of the fault, allowing technicians to check and take appropriate measures in a timely manner.

[0154] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method.

[0155] Figure 6 This is a schematic diagram of the structure of the server fan speed regulating device provided in the embodiment of the present application. Figure 6 As shown, the server fan speed regulating device 60 includes:

[0156] An acquisition module 601 is configured to acquire an initial fan speed value corresponding to a current air inlet temperature, an average power consumption of the first server in a previous cycle, and an average total power consumption of the first fan;

[0157] A determination module 602 is configured to determine a first energy efficiency value of the server in the previous cycle based on the first server average power consumption and the first fan average total power consumption, and trigger an efficiency speed regulation logic if the server's operating status in the previous cycle meets a preset operating status condition;

[0158] The performance speed control module 603 is used to:

[0159] Controlling each fan in the server to rotate according to a first fan speed value in a current cycle; the first fan speed value is less than an initial fan speed value;

[0160] Determine a second energy efficiency value of the server in the current period according to the second server average power consumption and the second fan average total power consumption in the current period;

[0161] The rotation speed of each fan is dynamically adjusted according to the first energy efficiency value and the second energy efficiency value.

[0162] In a possible implementation, the determining module 602 is further configured to:

[0163] Obtain the speed requirements of key components in the server;

[0164] Compare the speed demand value of the key component with the initial fan speed value to determine whether the key component triggered the proportional integral derivative speed regulation in the previous cycle;

[0165] If the key components do not trigger the proportional integral derivative speed regulation in the previous cycle, and the speed of each fan is adjusted according to the inlet air temperature speed regulation curve, it is determined that the working state of the server in the previous cycle meets the preset working state conditions.

[0166] In a possible implementation, the determining module 602 is specifically configured to:

[0167] Determine the difference between the first preset value and the energy efficiency ratio as the first energy efficiency value;

[0168] The energy efficiency ratio is the ratio of the average total power consumption of the first fan to the average power consumption of the first server.

[0169] In a possible implementation, the performance speed regulation module 603 is further configured to:

[0170] The difference between the initial fan speed value and the second preset value is determined as the first fan speed value.

[0171] In a possible implementation, the performance speed regulation module 603 is specifically configured to:

[0172] When the second energy efficiency value is greater than or equal to the first energy efficiency value:

[0173] Determine whether the key components trigger the proportional integral derivative speed control in the current cycle;

[0174] If the key component does not trigger the proportional-integral-differential speed regulation in the current cycle, and the power consumption fluctuation value of the key component is less than the preset fluctuation value, the difference between the first fan speed value and the second preset value is determined as the second fan speed value, and each fan is controlled to rotate according to the second fan speed value; or

[0175] If the key component triggers the proportional-integral-derivative speed regulation in the current cycle, and the power consumption fluctuation value of the key component is less than the preset fluctuation value, the efficiency speed regulation logic is interrupted, and the fan demand value corresponding to the key component is determined as the second fan speed value, and each fan is controlled to rotate according to the second fan speed value; or

[0176] If the key component does not trigger the proportional-integral-derivative speed regulation in the current cycle, and the power consumption fluctuation value of the key component is greater than or equal to the preset fluctuation value, the efficiency speed regulation logic is interrupted, and the fan speed value corresponding to the current air inlet temperature is determined as the second fan speed value, and each fan is controlled to rotate according to the second fan speed value; or,

[0177] If the key component triggers proportional integral differential speed regulation in the current cycle, and the power consumption fluctuation value of the key component is greater than or equal to the preset fluctuation value, the performance speed regulation logic is interrupted, and the maximum value of the fan demand value corresponding to the key component and the fan speed value corresponding to the inlet air temperature at the current moment is determined as the second fan speed value, and each fan is controlled to rotate according to the second fan speed value.

[0178] In a possible implementation, the performance speed regulation module 603 is specifically configured to:

[0179] When the second energy efficiency value is less than the first energy efficiency value, each fan is controlled to continue rotating according to the first fan speed value.

[0180] In a possible implementation, the server fan speed regulating device 60 further includes a fault determination module configured to:

[0181] When any fan fails, the power consumption value of the failed fan is compared with the power consumption values ​​of other fans except the failed fan among the multiple fans in the server;

[0182] If the power consumption value of the faulty fan is not 0 and the difference between the power consumption value of the faulty fan and any other fan is less than the preset difference, the fault is determined to be caused by abnormal frequency signal; or,

[0183] If the power consumption value of the faulty fan is not 0 and the difference between the power consumption values ​​of the other fans is greater than or equal to the preset difference, it is determined that the fault is caused by abnormal pulse width modulation signal; or,

[0184] If the power consumption of the faulty fan is 0, the fault is determined to be caused by an abnormal power supply signal.

[0185] In a possible implementation, the acquisition module 601 is specifically configured to:

[0186] Obtain the preset high-parameter inlet air temperature speed control curve;

[0187] Get the initial fan speed value corresponding to the current inlet air temperature from the inlet air temperature speed control curve.

[0188] For the description of the features in the embodiment corresponding to the server fan speed regulating device 60, reference can be made to the relevant description of the embodiment corresponding to the server fan speed regulating method, which will not be repeated here.

[0189] Figure 7 This is a schematic diagram of the structure of the baseboard management controller provided in this application. Figure 7 As shown, the baseboard management controller 10 provided in this embodiment includes: a processor 101 and a memory 102. Optionally, the baseboard management controller 10 further includes a communication component 103. The processor 101, the memory 102 and the communication component 103 are connected via a bus.

[0190] During the specific implementation process, the processor 101 executes the computer program stored in the memory 102, so that the processor 101 executes the above-mentioned server fan speed adjustment method embodiment.

[0191] The specific implementation process of the processor 101 can be found in the above method embodiment. Its implementation principle and technical effects are similar and will not be repeated here in this embodiment.

[0192] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the application may be directly executed by a hardware processor or by a combination of hardware and software modules within the processor.

[0193] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage.

[0194] A bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be divided into address buses, data buses, and control buses.

[0195] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. The computer program is configured to execute the steps of any of the above-mentioned server fan speed regulation method embodiments when running.

[0196] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0197] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in any of the above-mentioned server fan speed regulation method embodiments are implemented.

[0198] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of any of the above-mentioned server fan speed regulation method embodiments are implemented.

[0199] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0200] The above is a detailed introduction to a server cooling system, a server fan speed control method, and a baseboard management controller provided by the present application. This article uses specific examples to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core ideas of the present application. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of the present application.

Claims

1. A server cooling system, characterized in that: include: A baseboard management controller, a fan board, and a plurality of electronic fuse units; the fan board is provided with a plurality of fans, each of the fans is electrically connected to a corresponding electronic fuse unit, and each of the electronic fuse units is electrically connected to the baseboard management controller; The electronic fuse unit is used to collect the current value and voltage value of the fan at each sampling moment in the current cycle and send them to the baseboard management controller; The baseboard management controller is configured to determine an average total power consumption of the fans in the current cycle based on the current value and voltage value of each fan at each sampling moment in the current cycle; The baseboard management controller is further configured to, when the working state of the server in the previous cycle meets a preset working state condition, determine the first energy efficiency value of the server in the previous cycle based on the average power consumption of the first server and the average total power consumption of the first fan in the previous cycle, and trigger the efficiency speed regulation logic, wherein the preset working state condition is that the key components in the server in the previous cycle did not trigger the proportional integral differential speed regulation, and the speed of each fan in the server is controlled according to the inlet air temperature speed regulation curve; and determine the second energy efficiency value of the server in the current cycle based on the average power consumption of the second server and the average total power consumption of the second fan in the current cycle, wherein the energy efficiency value is determined based on the difference between the preset value and the energy efficiency ratio, and the energy efficiency ratio is the ratio of the average total power consumption of the first fan to the average power consumption of the first server; and when the second energy efficiency value is greater than or equal to the first energy efficiency value: determine whether the key components in the current cycle trigger the proportional integral differential speed regulation; If the key component does not trigger proportional-integral-differential speed regulation within the current cycle, and the power consumption fluctuation value of the key component is less than the preset fluctuation value, then the difference between the first fan speed value and the second preset value is determined as the second fan speed value, and each of the fans is controlled to rotate according to the second fan speed value, and the first fan speed value is less than the initial fan speed value; or If the key component triggers proportional-integral-derivative speed regulation in the current cycle, and the power consumption fluctuation value of the key component is less than the preset fluctuation value, the performance speed regulation logic is interrupted, and the fan demand value corresponding to the key component is determined as the second fan speed value, and each of the fans is controlled to rotate according to the second fan speed value; or If the key component does not trigger the proportional-integral-derivative speed regulation within the current cycle, and the power consumption fluctuation value of the key component is greater than or equal to the preset fluctuation value, then the performance speed regulation logic is interrupted, and the fan speed value corresponding to the current air inlet temperature is determined as the second fan speed value, and each of the fans is controlled to rotate according to the second fan speed value; or, If the key component triggers proportional-integral-derivative speed regulation in the current cycle, and the power consumption fluctuation value of the key component is greater than or equal to the preset fluctuation value, the performance speed regulation logic is interrupted, and the maximum value of the fan demand value corresponding to the key component and the fan speed value corresponding to the current inlet air temperature is determined as the second fan speed value, and each of the fans is controlled to rotate according to the second fan speed value; When the second energy efficiency value is less than the first energy efficiency value, each of the fans is controlled to continue rotating according to the first fan speed value.

2. The server cooling system according to claim 1, wherein: The baseboard management controller is further used for: When any of the fans fails, the power consumption value of the failed fan is compared with the power consumption values ​​of other fans among the plurality of fans except the failed fan; If the power consumption value of the faulty fan is not 0 and the difference between the power consumption value of the faulty fan and any other fan is less than a preset difference, the fault is determined to be caused by an abnormal frequency signal; or If the power consumption value of the faulty fan is not 0 and the differences between the power consumption values ​​of the other fans are greater than or equal to the preset differences, it is determined that the cause of the fault is an abnormal pulse width modulation signal; or, If the power consumption value of the faulty fan is 0, it is determined that the fault cause is an abnormal power supply signal.

3. The server cooling system according to claim 1, wherein: The baseboard management controller is further used for: When the speed of each fan is adjusted according to the energy efficiency values ​​of two adjacent cycles, a high-parameter inlet air temperature speed adjustment curve is used for speed adjustment.

4. A server fan speed adjustment method, characterized in that: include: Obtaining the initial fan speed value corresponding to the current inlet air temperature, the average power consumption of the first server in the previous period, and the average total power consumption of the first fan; If the working state of the server in the previous cycle meets a preset working state condition, a first energy efficiency value of the server in the previous cycle is determined based on the average power consumption of the first server and the average total power consumption of the first fan, and the efficiency speed regulation logic is triggered. The preset working state condition is that the key components in the server in the previous cycle did not trigger the proportional integral derivative speed regulation, and the speed of each fan in the server is adjusted according to the inlet air temperature speed regulation curve; The efficiency speed regulation logic includes: Controlling each fan in the server to rotate according to a first fan speed value in a current cycle; the first fan speed value is less than the initial fan speed value; determining a second energy efficiency value of the server in the current period based on the average power consumption of the second server and the average total power consumption of the second fan in the current period; the energy efficiency value is determined based on a difference between a preset value and an energy efficiency ratio, where the energy efficiency ratio is a ratio of the average total power consumption of the first fan to the average power consumption of the first server; When the second energy efficiency value is greater than or equal to the first energy efficiency value: determining whether the key component triggers proportional integral derivative speed regulation in the current cycle; If the key component does not trigger proportional-integral-differential speed regulation within the current cycle, and the power consumption fluctuation value of the key component is less than the preset fluctuation value, the difference between the first fan speed value and the second preset value is determined as the second fan speed value, and each of the fans is controlled to rotate according to the second fan speed value; or If the key component triggers proportional-integral-derivative speed regulation in the current cycle, and the power consumption fluctuation value of the key component is less than the preset fluctuation value, the performance speed regulation logic is interrupted, and the fan demand value corresponding to the key component is determined as the second fan speed value, and each of the fans is controlled to rotate according to the second fan speed value; or If the key component does not trigger the proportional-integral-derivative speed regulation within the current cycle, and the power consumption fluctuation value of the key component is greater than or equal to the preset fluctuation value, then the performance speed regulation logic is interrupted, and the fan speed value corresponding to the current air inlet temperature is determined as the second fan speed value, and each of the fans is controlled to rotate according to the second fan speed value; or, If the key component triggers proportional-integral-derivative speed regulation in the current cycle, and the power consumption fluctuation value of the key component is greater than or equal to the preset fluctuation value, the performance speed regulation logic is interrupted, and the maximum value of the fan demand value corresponding to the key component and the fan speed value corresponding to the current inlet air temperature is determined as the second fan speed value, and each of the fans is controlled to rotate according to the second fan speed value; When the second energy efficiency value is less than the first energy efficiency value, each of the fans is controlled to continue rotating according to the first fan speed value.

5. The server fan speed adjustment method according to claim 4, characterized in that: The method further comprises: Obtaining speed requirements of key components within the server; Comparing the speed demand value of the key component with the initial fan speed value to determine whether the key component triggers proportional integral derivative speed regulation in the previous cycle; If the key component does not trigger proportional integral derivative speed regulation in the previous cycle, and the speed of each fan is adjusted according to the inlet air temperature speed regulation curve, it is determined that the working status of the server in the previous cycle meets the preset working status conditions.

6. The server fan speed adjustment method according to claim 4, characterized in that: Before controlling each of the fans to rotate according to the first fan speed value in the current cycle, the method further includes: The difference between the initial fan speed value and the second preset value is determined as the first fan speed value.

7. The server fan speed adjustment method according to claim 4, characterized in that: The method further comprises: When any of the fans fails, the power consumption value of the failed fan is compared with the power consumption values ​​of other fans in the server except the failed fan; If the power consumption value of the faulty fan is not 0 and the difference between the power consumption value of the faulty fan and any other fan is less than a preset difference, the fault is determined to be caused by an abnormal frequency signal; or If the power consumption value of the faulty fan is not 0 and the difference between the power consumption values ​​of the other fans is greater than or equal to the preset difference, it is determined that the fault cause is an abnormal pulse width modulation signal; or, If the power consumption value of the faulty fan is 0, it is determined that the fault cause is an abnormal power supply signal.

8. The server fan speed adjustment method according to claim 4, characterized in that: The step of obtaining the initial fan speed value corresponding to the current air inlet temperature includes: Obtain the preset high-parameter inlet air temperature speed control curve; The initial fan speed value corresponding to the current inlet air temperature is obtained from the inlet air temperature speed control curve.

9. A baseboard management controller, characterized in that: include: Memory for storing computer programs; A processor is configured to implement the steps of the server fan speed adjustment method according to any one of claims 4 to 8 when executing the computer program.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of the server fan speed adjustment method according to any one of claims 4 to 8.

11. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the server fan speed adjustment method according to any one of claims 4 to 8 are implemented.

Images